A proposal for the reference intervals of the Italian microbiota "scaffold" in healthy adults

Abstract

Numerous factors, ranging from genetics, age, lifestyle, and dietary habits to local environments, contribute to the heterogeneity of the microbiota in humans. Understanding the variability of a "healthy microbiota" is a major challenge in scientific research. The gut microbiota profiles of 148 healthy Italian volunteers were examined by 16S rRNA gene sequencing to determine the range and diversity of taxonomic compositions in the gut microbiota of healthy populations. Possible driving factors were evaluated through a detailed anamnestic questionnaire. Microbiota reference intervals were also calculated. A "scaffold" of a healthy Italian gut microbiota composition was identified. Differences in relative quantitative ratios of microbiota composition were detected in two clusters: a bigger cluster (C₂), which included 124 subjects, was characterized by more people from the northern Italian regions, who habitually practised more physical activity and with fewer dietary restrictions. Species richness and diversity were significantly higher in this cluster (C₂) than in the other one (C₁) (C₁: 146.67 ± 43.67; C₂: 198.17 ± 48.47; F = 23.40; P < 0.001 and C₁: 16.88 ± 8.66; C₂: 35.01 ± 13.40; F = 40.50; P < 0.001, respectively). The main contribution of the present study was the identification of the existence of a primary healthy microbiological framework that is only marginally affected by variations. Taken together, our data help to contextualize studies on population-specific variations, including marginal aspects, in human microbiota composition. Such variations must be related to the primary framework of a healthy microbiota and providing this perspective could help scientists to better design experimental plans and develop strategies for precision tailored microbiota modulation.

Figure 1

Heatmap of the most represented phyla. Each column represents a single phylum, with each row representing a different subject/sample. Bray–Curtis distance was used as a clustering method. Two different clusters of subjects were selected as the best solution (see above).

Figure 2

Violin plots of the Firmicutes/Bacteroidetes ratio in the two clusters.

Figure 3

Spearman correlation plots of genera (cut off: relative abundance 0.5%) in the two clusters.